ComeçarComece de graça

Identification

In the following exercises you will apply to the Box-Jenkins methodology to go from an unknown dataset to a model which is ready to make forecasts.

You will be using a new time series. This is the personal savings as % of disposable income 1955-1979 in the US.

The first step of the Box-Jenkins methodology is Identification. In this exercise you will use the tools at your disposal to test whether this new time series is stationary.

The time series has been loaded in as a DataFrame savings and the adfuller() function has been imported.

Este exercício faz parte do curso

ARIMA Models in Python

Ver curso

Instruções do exercício

  • Plot the time series using the DataFrame's .plot() method.
  • Apply the Dicky-Fuller test to the 'savings' column of the savings DataFrame and assign the test outcome to result.
  • Print the Dicky-Fuller test statistics and the associated p-value.

Exercício interativo prático

Experimente este exercício completando este código de exemplo.

# Plot time series
____
plt.show()

# Run Dicky-Fuller test
result = ____

# Print test statistic
____

# Print p-value
____
Editar e executar o código